568 HENRY A. KOWLAND 



The substance d comes down slightly sooner than a by sulphate of 

 soda, but slightly slower by sulphate of potash. Hence, in purifying 

 yttrium (substance a) for the last time from the ce. group, sulphate of 

 potash will increase d in the filtrate and sulphate of soda will decrease it. 



Action of oxalic acid 



When the oxalates of the mixed earths, free from the ce. group, are 

 boiled in water to which nitric acid is added, they are more or less dis- 

 solved, leaving a coarse, heavy, red oxalate yielding a pale yellow oxide. 

 The nitrate, set aside to cool, deposits more of the oxalates and leaves a 

 filtrate which contains several of the unknown elements, as also what re- 

 mains of the ce. group. On separating the ce. group the remainder is 

 quite different from the heavy red oxalate, but there is far from complete 

 separation. The analysis showed the following: 



a, &, c, d, li, i, n. 



I have not found the separation particularly useful, and it seems to be 

 more apparent than real as tested by the spectroscope. 



Ferrocyanide of potassium 



This is the most useful process and easily separates the element a, 

 pure and free from all others. To obtain pure a from the mineral gado- 

 linite, Fergusonite or Samarskite: 



First obtain the crude mixed earths in the usual manner. Then sepa- 

 rate the cerium group as usual until the absorption bands of neodymium 

 no longer appear. For the complete separation without loss this must 

 be done several times, as much of the yttrium group is carried down 

 with the first precipitate, as we have before seen. 



The separation of the yttrium (a) from the other elements is effected 

 by precipitating the latter from a weak acid solution by ferrocyanide of 

 potassium. For this purpose the filtrate, after separating the cerium 

 group, can be used at once by slightly acidulating with nitric acid, dilut- 

 ing and adding a weak solution of ferrocyanide of potassium. No pre- 

 cipitate should appear at once, but by standing for an hour or so some 

 will come down. Add more ferrocyanide of potassium and repeat until 

 the filtrate no longer shows the bands of so-called erbium. After this 

 it is best to precipitate with oxalic acid or oxalate of potassium and 

 ignite the precipitate so as to get the earth. Dissolve this in nitric acid 

 and add only water enough to make a very concentrated syrupy solution. 



